By Nestor Lopez-Duran PhD
Is having multiple childcare arrangements during early childhood associated with increased behavioral problems in toddlers?
A review of Morrissey, T. (2009). Multiple Child-Care Arrangements and Young Children’s Behavioral Outcomes Child Development, 80 (1), 59-76 DOI: 10.1111/j.1467-8624.2008.01246.x
In the last issue of the journal Child Development, Dr. Taryn Morrissey conducts a comprehensive investigation of the effects of multiple childcare arrangements on childrens behavioral and emotional problems. Dr. Morrissey was particularly interested in one type of multiple childcare arrangements called arrangement multiplicity
. This refers to having multiple childcare providers at the same relative time, such as when a child is frequently looked after by a nanny, an aunt, and an older sister during the same week. This type of multiple arrangement is different than arrangement instability, such as when the child experiences frequent permanent changes in childcare (for example when the child changes daycare centers or nannies). The author wanted to examine only arrangement multiplicity because previous studies showing some potentially harmful effects of multiple arrangements have combined these two types of arrangements, so it is difficult to tell whether the potential harmful effects are due to having stable but multiple concurrent arrangements vs. having more long-term instability in childcare.
The author used data from the National Institute for Child Health and Human Development Study of Early Child Care and Youth Development to examine whether the number of non-parental concurrent child care arrangements was associated with adaptive social behaviors, internalizing problems (e.g., depression) and externalizing problems (e.g., aggression). The data included 1,229 families across the United States which were followed for up to 36 months after the childs birth. The author controlled for the question of stability by only including multiple arrangement that overlapped in time. Thus, if a parent reported using 2 sources of childcare but one had completely stopped before starting the other (such as permanently changing from one nanny to another), such arrangements were not considered to be concurrent.
The author found that 83% of children experienced at least one non-parental childcare arrangement by the age of 3, and 14% experienced at least 2 non-parental concurrent child care arrangements by this age. Children in multiple arrangements were found to have higher quality of primary care, higher instability of child care, and higher family income-to-needs ratio. Thus children with more concurrent arrangements seemed to have better daycare, families with more expending money, but more frequent changes in their daycare.
Higher number of concurrent arrangements were associated with lower levels of pro-social behaviors among boys and girls as reported by mothers, but this association was modest (effect size = -.21). Also as reported by mothers, higher number of concurrent arrangements were associated with higher disruptive behaviors but only among girls.
The surprising finding came when the author examined the reports of the non-parental caregivers. According to the non-parental caregivers, the age of the child appeared to play a major role. That is, higher number of concurrent arrangements were not associated with disruptive behaviors among the 2-year-old kids. However, among the 3-year-olds, higher number of concurrent arrangements were associated with LESS disruptive behaviors. Concerning externalizing and internalizing problems, multiple arrangements were associated with more problems, but only in the 2-year-olds. By age 3, multiple arrangements were unrelated to internalizing or externalizing problems.
The author concludes that multiple arrangements appear to be associated with modest increases in behavior problems especially among girls, but only when reported by the mothers. When reported by the non-parental caregiver, the association between multiple childcare and behavior problems are only observed in younger children. Who is right? It depends who you ask. This issue of report bias is highly controversial. Some argue that maternal reports reflect how the mother perceives the child but not necessarily how the child truly behaves, especially when compared to his/her peers. For example, a mother may think that her child is an angel when the child is actually very disruptive, or alternatively, a mother may think her child is very bad, when the child is actually much better than most of his/her peers. It is also possible that the teachers and caregivers simply have a wider frame of reference since they are exposed to many more children on a daily basis, so they are making a judgment about the childs behavior in the context of how other kids behave. However, it can also be argued that mothers spend significantly more one-to-one time with their children, so that they are more likely to see behavior problems that the teacher/caregiver may miss. However, my colleagues at the University of Michigan have argued that parents and teachers are both correct, in that their reports are based on different but highly relevant contexts (see Kerr et al. DOI 10.1111/j.1469-7610.2007.01776.x).
In sum, the study suggests that multiple childcare arrangements may be associated with higher levels of behavior problems, but there is some discrepancy between mothers and other caregivers. However, the author also correctly discusses the issue of “causality”. That is, this type of study can not conclusively state that multiple arrangements cause higher levels of behavior problems. An alternative explanation is that children with more behavior problems are more likely to elicit or require multiple arrangements, or that multiple arrangements is associated with a factor, not examined in this study, that may be the cause of the increases in problems (see below). Yet, the results are consistent with current developmental theories of the importance of stability in multiple domains (family, schools, peers, etc) during early childhood.
A final, and
a bit more technical, issue. The author was interested in using a measure of concurrent multiplicity that was not confounded by instability of child-care (frequent permanent changes). However, the number of non-overlapping frequent changes (a possible index of instability) was not used in their statistical models as a control variable. This limitation means that the results can still be a byproduct of instability. That is, families with multiple concurrent arrangements may also be the families with high levels of unstable childcare, and it may be this instability (and not the concurrent multiplicity) that is associated with behavior problems.